A Method of Channel Capacity Optimization Based on Dynamically Adjusted Inertia Weight Acceleration Factor in Cognitive Sensing Network

Abstract

The optimization of channel capacity in cognitive sensor networks is a complicated optimization problem. The traditional gradient search method based on the analysis has more restrictions on the objective function, and high complexity, and can not determine the convergence. Aiming at the inherent problems of the traditional gradient search algorithm, the particle swarm optimization(PSO) with simple and easy to implement, distributed computing and fast convergence speed can be used to solve the problem of channel capacity optimization. It is difficult to balance the global search with the local search by adopting a standard particle swarm algorithm with fixed algorithm parameters, which can not solve the premature convergence problem that may occur. The specific meaning of each parameter of the algorithm is analyzed in this paper, and an improved particle swarm optimization algorithm based on dynamic adjustment of inertia weight acceleration factor(DWAPSO) is proposed, and the improved particle swarm optimization algorithm is applied to the optimization of channel capacity in cognitive sensor networks. The simulation results show that the improved channel capacity optimization algorithm(DWAPSO-CA) can speed up the convergence rate, increase the system capacity and get a lower bit error rate.